Effect estimates can be accurately calculated with data digitally extracted from interrupted time series graphs

Turner, Simon; Korevaar, Elizabeth; Cumpston, Miranda; Kanukula, Raju; Forbes, Andrew Benjamin; McKenzie, Joanne E.

doi:10.1101/2022.09.12.22279878

Cited by 4 publications

(6 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This may require re‐analysis of the ITS studies prior to their inclusion in meta‐analysis. ITS data are often available in figures in publications, 12 from which data can be digitally extracted and effect estimates accurately calculated 64 . However, caution is required in relying on the estimated heterogeneity when there are few ITS studies and the ITS have few datapoints.…”

Section: Discussionmentioning

confidence: 99%

Evaluation of statistical methods used to meta‐analyse results from interrupted time series studies: A simulation study

Korevaar,

Turner,

Forbes

et al. 2023

Research Synthesis Methods

Self Cite

View full text Add to dashboard Cite

Interrupted time series (ITS) are often meta‐analysed to inform public health and policy decisions but examination of the statistical methods for ITS analysis and meta‐analysis in this context is limited. We simulated meta‐analyses of ITS studies with continuous outcome data, analysed the studies using segmented linear regression with two estimation methods [ordinary least squares (OLS) and restricted maximum likelihood (REML)], and meta‐analysed the immediate level‐ and slope‐change effect estimates using fixed‐effect and (multiple) random‐effects meta‐analysis methods. Simulation design parameters included varying series length; magnitude of lag‐1 autocorrelation; magnitude of level‐ and slope‐changes; number of included studies; and, effect size heterogeneity. All meta‐analysis methods yielded unbiased estimates of the interruption effects. All random effects meta‐analysis methods yielded coverage close to the nominal level, irrespective of the ITS analysis method used and other design parameters. However, heterogeneity was frequently overestimated in scenarios where the ITS study standard errors were underestimated, which occurred for short series or when the ITS analysis method did not appropriately account for autocorrelation. The performance of meta‐analysis methods depends on the design and analysis of the included ITS studies. Although all random effects methods performed well in terms of coverage, irrespective of the ITS analysis method, we recommend the use of effect estimates calculated from ITS methods that adjust for autocorrelation when possible. Doing so will likely to lead to more accurate estimates of the heterogeneity variance.

show abstract

Section: Discussionmentioning

confidence: 99%

Evaluation of statistical methods used to meta‐analyse results from interrupted time series studies: A simulation study

Korevaar,

Turner,

Forbes

et al. 2023

Research Synthesis Methods

Self Cite

View full text Add to dashboard Cite

show abstract

“…Digitally extracted time series data from published gures in the review using WebPlotDigitizer 29 . This data extraction tool has been shown to yield data that can be used to obtain accurate estimates of the effect estimates and standard errors from published ITS graphs 30 .…”

Section: Methods To Obtain Time Series Datamentioning

confidence: 99%

Comparison of statistical methods used to meta-analyse results from interrupted time series studies: an empirical study

Korevaar

Turner

Forbes

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Background The Interrupted Time Series (ITS) is a robust design for evaluating public health and policy interventions or exposures when randomisation is infeasible. Several statistical methods are available for the analysis and meta-analysis of ITS studies. We sought to empirically compare available methods when applied to real-world ITS data. Methods We sourced ITS data from published meta-analyses to create an online data repository. Each dataset was re-analysed using two ITS estimation methods. The level- and slope-change effect estimates (and standard errors) were calculated and combined using fixed-effect and four random-effects meta-analysis methods. We examined differences in meta-analytic level- and slope-change estimates, their 95% confidence intervals, p-values, and estimates of heterogeneity across the statistical methods. Results Of 40 eligible meta-analyses, data from 17 meta-analyses including 283 ITS studies were obtained and analysed. We found that on average, the meta-analytic effect estimates, their standard errors and between-study variances were not sensitive to meta-analysis method choice, irrespective of the ITS analysis method. However, confidence interval widths and p-values for the meta-analytic effect estimates varied depending on the choice of confidence interval method and ITS analysis method. Conclusions The meta-analysis effect estimates, their standard errors and between-study variance estimates were minimally impacted by ITS analysis and meta-analysis method choice. However, the confidence interval widths and p-values could vary according to the statistical method, which may impact interpretations and conclusions of a meta-analysis. This empirical study, in conjunction with evidence from numerical simulation, allows for a more complete understanding of which methods should be used in different scenarios.

show abstract

“…A standardised data collection form was used, including the following fields: title, authors, year of publication, study design, period and setting, sample size, inclusion and exclusion criteria, method of recruitment, participant characteristics, intervention details, control group descriptions, outcome measures, assessment time points and study results. WebPlotDigitizer V.4.6 was used to retrieve data when data were reported in graphical format 43 44…”

Section: Methodsmentioning

confidence: 99%

Effectiveness of social prescribing for chronic disease prevention in adults: a systematic review and meta-analysis of randomised controlled trials

Htun

Teshale

Cumpston

et al. 2023

J Epidemiol Community Health

Self Cite

View full text Add to dashboard Cite

BackgroundSocial prescribing (SP) enables healthcare professionals to link patients with non-medical interventions available in the community to address underlying socioeconomic and behavioural determinants. We synthesised the evidence to understand the effectiveness of SP for chronic disease prevention.MethodsA systematic literature search was conducted using five databases and two registries. Eligible studies included randomised controlled trials of SP among community-dwelling adults recruited from primary care or community setting, investigating any chronic disease risk factors defined by the WHO (behavioural factors: smoking, physical inactivity, unhealthy diet and excessive alcohol consumption; metabolic factors: raised blood pressure, overweight/obesity, hyperlipidaemia and hyperglycaemia). Random effect meta-analyses were performed at two time points: completion of intervention and follow-up after trial.ResultsWe identified nine reports from eight trials totalling 4621 participants. All studies evaluated SP exercise interventions which were highly heterogeneous regarding the content, duration, frequency and length of follow-up. Majority of studies had some concerns for risk of bias. Meta-analysis revealed that SP likely increased physical activity (completion: mean difference (MD) 21 min/week, 95% CI 3 to 39, I2=0%; follow-up ≤12 months: MD 19 min/week, 95% CI 8 to 29, I2=0%). However, SP may not improve markers of adiposity, blood pressure, glucose and serum lipid. There were no eligible studies that primarily target unhealthy diet, smoking and excessive alcohol drinking behaviours.ConclusionsSP exercise interventions probably increased physical activity slightly; however, no benefits were observed for metabolic factors. Determining whether SP is effective in modifying the determinants of chronic diseases and promotes sustainable healthy behaviours is limited by the current evidence of quantification and uncertainty, warranting further rigorous studies.PROSPERO registration numberCRD42022346687.

show abstract

Effect estimates can be accurately calculated with data digitally extracted from interrupted time series graphs

Cited by 4 publications

References 31 publications

Evaluation of statistical methods used to meta‐analyse results from interrupted time series studies: A simulation study

Evaluation of statistical methods used to meta‐analyse results from interrupted time series studies: A simulation study

Comparison of statistical methods used to meta-analyse results from interrupted time series studies: an empirical study

Effectiveness of social prescribing for chronic disease prevention in adults: a systematic review and meta-analysis of randomised controlled trials

Contact Info

Product

Resources

About